Energy market reform options in Iceland - Copenhagen Economics · Policy area 3: Integrated assessment of energy and environmental objectives Harnessing power from natural resources

Energy market reform options in Iceland Promoting security of supply and natural resource value

20 February 2017

Energy market reform options in Iceland

Promoting security of supply and natural

resource value

Authors:

Helge Sigurd Næss-Schmidt, Partner

Martin Bo Westh Hansen, Managing Economist

David von Below, Economist

Energy market reform

options in Iceland

Promoting security of

supply and natural

resource value

Preface

Iceland is a relatively small and isolated region with abundant access to cheap natural en-

ergy resources. While historically this resource has not been in high demand, this has now

changed. Numerous power intensive consumers are to an increasing degree seeing Ice-

land as an attractive location for their facilities. While this is a positive situation for Ice-

land, it also raises some questions, such as:

What will it imply for Icelandic households in terms of security of supply and implica-

tions for the energy bill?

Does Iceland have the best framework in place for obtaining value from its natural re-

sources?

How can the value be increased without causing irreversible damage to the environ-

ment?

Against this background, Copenhagen Economics has been asked by Landsvirkjun to take

stock of the Icelandic energy market regulation and how this matches the new develop-

ments for Iceland. In this study, we find that there are in fact a number of key challenges

which should be addressed, and suggest potential options for reform.



resource value

Table of contents

Preface 0

Executive summary 4

1 Taking stock of the Icelandic energy market and challenges going forward 9

1.1 Threats to security of supply for households and SMEs 9

1.2 The power sector can deliver more value to Iceland 13

2 Five options for policy reform 21

2.1 Enhancing security of supply for households and SMEs 21

2.2 Options to ensure fair value to the Icelandic community

from its energy resources 22

2.3 Integrating value from environment and energy 24

2.4 Market organisation in the Icelandic energy markets 24

2.5 Interconnector project to be considered as an option 30

References 31



resource value

List of tables

Table 1 Market share of the largest power generator across

European countries ...........................................................................24

Table 2 Landsvirkjun’s generation is low in international

comparison ....................................................................................... 25

Table 3 Very few small isolated power markets with exchange-

based trade ....................................................................................... 28



resource value

List of figures

Figure 1 The buffer of spare capacity is shrinking ............................ 10

Figure 2 Industrial power prices are approaching

household/SME prices ...................................................................... 11

Figure 3 Costs increase with power supplied by diesel

generators ......................................................................................... 13

Figure 4 The price of power to power-intensive industry has

historically been low in Iceland ........................................................ 14

Figure 5 Low return on capital in Icelandic power sector ................. 15

Figure 6 Potential resource rent increases with higher power

prices ................................................................................................. 16

Figure 7 Trade intensity of aluminium is comparably high .............. 17

Figure 8 Illustration of room for manoeuvre in price setting ........... 18

Figure 9 Potential new consumers who could consider Iceland ....... 19

Figure 10 Master Plan restrictions increases power generation

costs ................................................................................................. 20



resource value

List of boxes

Box 1 The Electricity Pool of England and Wales..............................29



resource value

4

Executive summary

Iceland has access to abundant resources of cheap, carbon free energy based mainly on

geothermal and hydro sources. However, as Iceland is isolated from the international elec-

tricity market, demand for these resources has been low. Consequently, it has historically

been challenging to turn these resources into value for the Icelandic community, other than

by offering energy at low costs to households and businesses. The first step taken in this

direction was to attract international power intensive consumers to Iceland by offering

power at low prices, in combination with very stable conditions and risk hedging contracts.

From the late 2000s onward, Iceland has seen a large increase in power demand and a

broader base of industrial consumers such as data centres and new metal industries. Fol-

lowing the introduction of the new Energy Law in 2003 and combined with recent renewals

of previous long-term power contracts, this has brought to the fore questions about whether

Iceland is getting a fair value out of its natural resources, and if the regulatory framework

in place is the best for addressing Iceland’s particular energy market characteristics.

In this study, we argue that Iceland has already taken appropriate steps towards achieving

fair value, but that there is potential for further improvement particularly in two areas.

Based on our assessment of the situation in Iceland, we have also identified five areas for

potential policy reform, which can combine into a policy package improving the overall en-

ergy market in Iceland to the benefit of the entire community.

Challenge 1: Security of supply for households and SMEs needs addressing

There are several indications suggesting that security of supply for the Icelandic households

and small- and medium-sized enterprises (SMEs) is deteriorating. The existing capacity

buffer is decreasing and is already almost exhausted. Moreover, due to increased demand

from industrial consumers, the historical price premium obtained from supplying to house-

holds/SMEs, rather than to the power-intensive industry, is shrinking. This makes it rela-

tively more attractive to supply to the power intensive segment instead of to house-

holds/SMEs. Indeed, several energy companies have publicly announced the possibility of

withdrawal of capacity from the household/SME sector to cater for the power intensive

segment.

In addition, the institutional responsibilities for monitoring and ensuring security of supply

are not clearly defined, and there are limited signals to investors to build new capacity via

the price. This development is of concern to households and businesses in Iceland, who

might need to look towards local back-up capacity such as expensive and polluting diesel

generators. Action is needed to address the situation.



resource value

5

Challenge 2: Potential to derive higher value from energy resources while pre-

serving other important objectives

Recent negotiations with power intensive industry has aimed for a better balance between

obtaining revenue to the Icelandic community while not jeopardising its position as an at-

tractive location for international power-intensive consumers. Indeed, our analysis sug-

gests that the historical revenue has been too low, with average return on capital on Ice-

landic power plants being significantly below international benchmarks, even with very lit-

tle revenue collected from ‘rent like’ taxes.

Obtaining more value through higher electricity prices is likely to benefit the Icelandic com-

munity strongly, even if it also implies increased prices for households. We find that in-

creasing the current average price of existing power contracts to an international bench-

mark such as the Nordpool price could bring additional value of USD 127–555 million per

year to the Icelandic community, depending on the future developments in the Nordpool

price.

Price increases must be seen in the context of maintaining Iceland as an attractive location

for international power intensive consumers. This applies both to deals with new consum-

ers, and to deals with existing consumers where relatively stable framework conditions over

time are important for investment attractiveness. As long as the conditions offered in Ice-

land are at least as good as in alternative locations, e.g. by offering power prices against an

international benchmark price and with contracts that are formed and honoured in a cred-

ible manner, prices can increase without undermining Iceland’s position as an attractive

location.

If it proves economical to expand power generation, it is important that this is done with

an integrated view of the value to Iceland both from power generation and from environ-

mental appreciation.

Based on our analysis, we have identified five areas for policy reform which could constitute

a strong reform package for the future energy policy of Iceland.

Policy area 1: Options for improving security of supply to households/SMEs

A number of options should be considered in order to address the risk of insufficient secu-

rity of supply for households and SMEs:

Firstly, the market signal of scarcity to investors could be improved. If the price of electric-

ity does not increase as a reaction to a lack of supply, investors will not face incentives to

invest in new capacity.

Secondly, the institutional responsibility for ensuring security of supply should be further

defined. The current lack of clarity leaves the potential problem unmonitored. In addition,

market participants have highlighted that in the absence of a clear institutional responsi-

bility, Landsvirkjun is implicitly expected to have sufficient capacity at its disposal. This is

not an appropriate role for an energy company in a competitive environment.



resource value

6

Thirdly, in the absence of a price increase in the household sector, subsidies for supplying

to this sector could be considered, thereby giving energy companies an economic incentive

to continue to do so. There are several potential models for achieving this, for example an

explicit subsidy (capacity payment) or an implicit subsidy by requiring energy producers to

supply to this segment. Such models should be assessed with respect to EU State Aid rules.

Policy area 2: Options for ensuring fair value to Iceland from power genera-

tion

The current setup for extracting value to the Icelandic community is to a large extent left to

the discretion of Landsvirkjun, in the absence of clear instructions from its owner, the Ice-

landic state. As Landsvirkjun is also a player in a competitive market, this leaves the role

and objectives of Landsvirkjun unclear. It may also run the risk of inhibiting the degree of

competition between energy producers if Landsvirkjun’s required return on equity is lower

than the return on equity of commercially owned producers.

Instead, the Icelandic community should consider defining its expected return from power

generation. The return should be as high as possible while still rendering Iceland an attrac-

tive location for electricity intensive consumers; consequently moving from cost-based

pricing to opportunity cost pricing. An anchor for this could be to aim at prices close to

international benchmark prices for power.

Two options to achieve this could be 1) setting up a transparent rent tax system to ensure

that the Icelandic community would reap a fair value from its natural resources, and 2)

defining a required return on equity for state owned companies in excess of the rent tax,

thereby ensuring competition on equal terms with commercial entities.

Several different designs of the rent tax could be considered, e.g. a tax element linked to an

international benchmark,1 thereby mimicking the attractiveness of locating power inten-

sive consumers in Iceland vis-à-vis other international locations. A national rent tax would

also have the benefit of potentially replacing different and non-transparent local rent-like

taxes levied by municipalities such as, e.g. licensing/concession fees. Such intervention

should consider the impact on local municipalities and e.g. be accompanied by State com-

pensation to municipalities.

Policy area 3: Integrated assessment of energy and environmental objectives

Harnessing power from natural resources can often be in opposition to enjoying the envi-

ronmental value of the same natural resources. In Iceland, this trade-off is currently ac-

tively assessed in the Master Plan act, where specific energy projects are deemed to be in

accordance or not with environmental objectives. This type of framework is commendable,

as it has the potential to make an integrated assessment taking into account multiple ob-

jectives.

However, concerns have been raised that the practical assessment has not sufficiently taken

into account the value from energy generation, thereby risking suboptimal outcomes for

1 For example the Nordpool spot price



resource value

7

the Icelandic community. Specifically, when restricting particular zones or power expan-

sion projects, the economic cost of doing so in terms of increased cost of energy generation,

or even the resulting ‘too little energy production’, should be included. This becomes in-

creasingly important if the demand for power increases, as power generation will then po-

tentially generate a higher value to the Icelandic community. The cost to Iceland from re-

stricting this will then also increase.

Policy area 4: Market organisation seem broadly appropriate but options

could be considered especially in the household market

In the recent Icelandic policy debate, concerns have been raised that the degree of market

concentration in power generation could potentially give rise to abuse of monopoly power.

The particular concern seems to be about the risk of prices being ‘too high’. We argue that

the concern seems misplaced. We base this on a number of arguments, also suggesting that

prices in Iceland are more likely to be ‘too low’ than ‘too high’:

Firstly, while Landsvirkjun holds an important position among energy producers in Ice-

land (about 70 per cent overall, and about 55 per cent of power sold to the retail market),

this is not uncommon when compared internationally. Indeed in 14 out of 24 EU countries

the incumbent electricity generator holds a market share of more than 50 per cent, and in

six countries the market share is above 80 per cent. Moreover, the size of the larger energy

companies in Europe is significantly higher than Landsvirkjun,2 suggesting that while

Landsvirkjun is a large player in Iceland, economies of scale suggest that it could be even

bigger.

Secondly, the market for selling power to international, energy-intensive consumers has a

very competitive nature. This implies that – no matter the market concentration in Iceland

– power prices cannot exceed international benchmarks as the power-intensive consumers

can choose to establish production somewhere else. Instead, it is likely to be an advantage

for Iceland to have a strong counterpart to the large international power consumers in con-

tract negotiations.

Thirdly, selling power at low prices will not improve market efficiency.3 As long as a domi-

nant player in the market does not restrict capacity utilisation, there is no rationale for low

prices as such, as it will not distort market decisions.

Fourthly, it is important that there are proper incentives to invest in new capacity, and that

a dominant player cannot restrict the expansion of potentially profitable new capacity. This

is a case for ensuring that prices are not being kept artificially low to prevent new market

entry, but instead being flexible enough to rise in order to make new investments viable.

Fifthly, there is currently ample room for other energy companies to invest in new capacity,

both in small-scale assets such as wind turbines and small hydro facilities, and in larger

2 Vattenfall for example, generates about 15 times more energy than Landsvirkjun, and the largest companies such as E.ON

and EDF more than 50 times more. 3 Which was otherwise suggested in a recent report, Christensen (2016)



resource value

8

assets. This is indeed taking place currently, and potential restrictions are more likely to be

found in environmental concerns than in abuse of market position.

An area where attention could be paid, however, is the risk of operational inefficiency of

the dominant player Landsvirkjun. In the absence of strong competitive pressure, a domi-

nant player may develop inefficiencies in terms of, e.g., too high salaries and benefits, and

lack of innovation. While we have not encountered reasons to believe that such inefficien-

cies currently exist, a rethink of the governance structure could be considered. Clarity could

be introduced by defining the expected contribution to the state through a rent tax model

and a required return on equity, as also suggested above.

Previously, suggestions have been made about making the wholesale market more ‘market

based’, e.g., by establishing a power exchange. While exchange based trading has a number

of desirable properties, such as lowering transaction costs for market participants, and giv-

ing rise to a price reflecting the true valuation of power, it requires a significant organisa-

tional setup to function, and the establishment of a well-functioning financial market. The

Icelandic retail market is quite small, and a fully-fledged exchange solution might struggle

to provide benefits in excess of the costs of administration. In addition, a market based

almost purely on energy sources with zero (or very low) marginal costs may prove difficult

to run without high price fluctuations and uncertainty.

Another option could be inspired by the so called ‘Energy pool’ in the UK in 1990. The main

idea behind such a pool is to introduce an independent buying agent on behalf of retail

consumers (not power intensive industry). This agent would thus be responsible for pur-

chasing sufficient energy from generators at the best possible price. Such a model would

likely achieve both security of supply in the segment, and a market-based pricing of energy.

Such a model would require economic regulation of the buying agent in order to determine

a cost-reflective retail price. The model would also imply unbundling between the genera-

tion and retail arm for the local energy companies.

The Icelandic TSO Landsnet is currently owned by the Icelandic energy companies, in par-

ticular Landsvirkjun. This may create improper incentives in the choices made by

Landsnet, e.g. related to investments in new transmission capacity, including connections

to new power plants. This is not ideal and a plan for divestment could be considered.

Policy area 5: Maintain an interconnector as an option

There are numerous reasons why an interconnector to, e.g., the UK could be an attractive

investment for Iceland, both in terms of security of supply and value creation for Iceland.

This is a long-term option, as lead times on decisions and subsequent construction are sub-

stantial. In the meantime, it is important that potential policy reform is designed with a

view to this as a possible long-term solution, e.g., by not locking Iceland into structures that

will prove uneconomic in the case that an interconnector is built. One example could be to

refrain from engaging in very long-term power contracts, unless they are linked to the price

that would occur after building an interconnector.



resource value

9

Chapter 1

1 Taking stock of the Icelandic energy market and challenges going forward

Iceland has access to abundant natural resources of cheap, carbon-free geothermal, hydro

and wind energy. These resources have historically been turned into value to the Icelandic

community by providing energy at low prices to households and businesses, as well as by

attracting international power intensive consumers to Iceland by offering low energy

prices, in combination with very stable conditions and risk hedging features.

In recent years, Iceland has seen a large increase in electricity consumption and a broader

base of industrial consumers such as data centres and new metal industries. Together with

recent renewals of existing long-term contracts, this has brought to the fore questions of

whether Iceland is obtaining a fair value out of its natural resources, and if the regulatory

framework in place is the best for addressing Iceland’s particular energy market character-

istics.

We have identified two broad areas with potential challenges for the Icelandic community:

1) security of supply to households and SMEs is deteriorating and 2) the power sector has

not delivered enough value to the Icelandic community. In this chapter, we describe these

challenges, and set the scene for the policy reform discussion in Chapter 2.

1.1 Threats to security of supply for households and SMEs Iceland’s power generation capacity has historically been higher than demand by some

margin, but this buffer has been shrinking, due to increasing demand for power. Lands-

virkjun’s energy margin, i.e., the difference between yearly contractible firm power4 and

sold power, has experienced a reduction from around 4 per cent of capacity (corresponding

to around 500 GWh/year) in 2008 to around zero in 2015, see Figure 1. The negative ca-

pacity margin already in 2015 is the result of relatively high precipitation, which meant that

Landsvirkjun could sell more than just its ‘contractible firm power’.

Importantly, given the expected demand growth just from households and SMEs (not tak-

ing into account power intensive industry), the energy margin will be significantly reduced

in the coming years in the absence of new capacity coming online. Without new capacity, it

is estimated that the capacity deficit could be approaching 4 per cent of capacity by 2020,

see Figure 1. If local energy companies increasingly channel their energy from households

to power intensive industry, the overall Icelandic capacity deficit will be even worse.

4 Contractible firm power refers to the guaranteed power that Landsvirkjun can produce in a year. This is based on a low-

precipitation year, with a corresponding low production of hydropower. The amount of power that Landsvirkjun can sell in

a year fluctuates naturally with the inflow of water into the hydropower system.



resource value

10

Figure 1 The buffer of spare capacity is shrinking

Note: The figure shows Landsvirkjun’s energy margin, i.e., unutilised capacity that could be activated, in the

period 2008–2015. The dashed line gives the projected energy margin until 2020, given the expected

demand increase in the retail segment (not including power intensive consumers).

Source: Copenhagen Economics based on Landsvirkjun and Orkustofnun (2015)

Three concerns for the security of supply for households and SMEs are

forming on the horizon

This shrinking capacity margin is cause for concern for the security of supply for house-

holds and SMEs5, due to a combination of three factors: the lack of a price signal in the

wholesale segment, the lack of a clear responsibility to ensure security of supply, and the

long lead times to expand generation capacity.

Firstly, there is limited market signal in the price formation in this segment – potential

investors cannot form an expectation of likely scarcity. Until now, the security of supply in

the wholesale segment has been ‘protected’ by relatively high prices compared to power

intensive consumers, giving an incentive for producers to cater for this segment.6 However,

in light of recent increases in demand, and considering that recently closed power contracts

with power-intensive consumers have involved relatively high prices, this price difference

is likely to be reduced, see Figure 2.

5 It is not a cause of concern for the power intensive companies, as generators are contractually obliged to supply power

through PPAs 6 It is important to note that the two segments differ with respect to the cost of supplying power. A generator can supply to an

industrial consumer at a lower cost than the wholesale market, due to the very stable and predictable loads in the PPAs

-4%

0%

4%

2010 2015 2020

% of total

generation capacity



resource value

11

Figure 2 Industrial power prices are approaching house-

hold/SME prices

Note: Constant (2016) prices.

Source: Copenhagen Economics based on Landsvirkjun

As industrial prices approach the prices charged to households and SMEs, there is less in-

centive for generators to supply to the latter segment. Indeed, some producers have already

announced that they are considering to sell more to the industrial segment at the expense

of the household segment.

In a competitive market, this situation would put an upwards pressure on the prices in the

wholesale segment, and lead to higher investments. However, the wholesale price is to a

large extent determined by Landsvirkjun, rather than in a competitive manner. As a large

state owned energy company, the pricing decision may to some extent be implicitly influ-

enced by political resistance to increasing the energy bill of households and SMEs. This

may hold back the natural upward adjustment of the price, and as such there is a risk that

the household/SME segment will increasingly be at risk of power shortages.

Secondly, the role and responsibility of ensuring security of supply is not clearly defined.

The lack of a price signal described above would be less of a problem if there were a clear

responsibility for ensuring that there is power available to households and SMEs, but this

does not seem to be the case. The Electricity Act (2003) mentions security of supply, mainly

in the context that the TSO shall maintain and develop the transmission system in a way

that takes into account security of supply. Article 9 of the Act further stipulates that the

TSO shall ensure the security and quality of the electricity delivery. However, this relates

to balancing of supply and demand, e.g. by ensuring an adequate supply of spinning re-

serves, rather than the kind of long-term security of supply concern under discussion here.

2

3

4

5

6

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2020 2025 2030 2035

ISK/kWh

Wholesale

Power intensive clients



resource value

12

Thirdly, expansions of generation capacity are subject to long lead times. The issues laid

out above would be less severe if new capacity could be brought in quickly. However, once

a problem of scarce capacity has been identified, the lead time to the new generation being

in place is significant. For projects already approved in the Master Plan, the lead time is

likely to be around four years for geothermal and hydro projects, and around two years in

the case of smaller wind projects. The time required to develop a completely new project is

substantially longer, perhaps around ten years. In this period, supply will be seriously lim-

ited, or only possible using expensive, inefficient and polluting technologies such as local

diesel generators.

A high security of supply is crucial for smaller power consumers

SMEs in Iceland are very dependent on a stable supply of electricity. Power shortages would

imply severe problems, and in the event that shortages start to become a serious issue,

businesses are likely to secure their own supply of power with, e.g., local diesel generators.

As an example, consider a fish processing plant. Having to invest in local generation capac-

ity through, e.g., diesel generators would give rise to electricity cost increases of a magni-

tude of around 200 USD/MWh or more (a 400% increase), and will in addition lead to

emissions of CO2 and other pollutants.7

If a fish processing plant had to cover 25% of its electricity consumption with a local diesel

generator, this would imply a rise in electricity costs of 60–90%, and a rise in total costs of

1.2–1.8%. In an extreme case where all power must be generated locally with a diesel gen-

erator, the total cost increase is on the order of 5–7%, thereby reducing the international

competitiveness of the sector, see Figure 3.

7 Based on levelised cost of energy of investing in a new diesel generator, based on Lazard Levelised cost of storage analysis

(2015)



resource value

13

Figure 3 Costs increase with power supplied by diesel genera-

tors

Note: Based on the cost structure of an average fish processing plant in Iceland, and on the levelized cost of

energy for diesel generators. The low end of the range represents continuous operation of the genera-

tor; high end represents intermittent operation of the generator.

Source: Statistics Iceland (Operating accounts of fish processing 1997-2014); Landsvirkjun; Lazard Levelized

cost of storage analysis (2015)

1.2 The power sector can deliver more value to Iceland Iceland has an abundance of natural resources, which can provide value to the Icelandic

community in several ways. One way is for Icelanders to enjoy the natural environment,

and another is to use the natural resources to create jobs and economic activity by providing

power to industry or by promoting tourism in the Icelandic nature. In this section, we argue

that current regulatory framework in place can be improved to derive more value from

power generation without jeopardising environmental value.

When discussing if Iceland is deriving a fair value from its natural resources, it is important

to define ‘value’. Traditionally, revenue from the power sector has been seen as less im-

portant since value has been generated through down-stream production of goods, most

predominantly aluminium. However, basic indicators of value created in power-intensive

industry, such as the share of GDP (6–7 per cent) and exports (40 per cent), do not neces-

sarily reflect value to the Icelandic people. What is important is the added value generated

and kept in Iceland.

The power sector has historically not generated the value that could be

expected given its potential

Power prices to power-intensive industrial consumers in Iceland have historically been low

in international comparison. The price of power in Iceland has generally been substantially

lower than in the other Nordic countries, which in turn have relatively low power prices by

0%

2%

4%

6%

8%

25% 50% 75% 100%

Cost increase

Proportion of power from diesel generators



resource value

14

European standards – indeed, the UK has significantly higher prices, especially in the most

recent years, see Figure 4.8

Figure 4 The price of power to power-intensive industry has his-

torically been low in Iceland

Note: The figure shows the cost of power only, excluding transmission/distribution costs and taxes. For Ice-

land, the prices are average prices on PPA contracts sold by Landsvirkjun also in the past, and there-

fore this does not reflect the price of a new power contract in a given year.

Source: Eurostat (nrg_pc_205_c); Statistics Norway; Landsvirkjun; Central Bank of Iceland

Another way to illustrate the value delivered to the Icelandic community from its power

production is to look at the return on capital in the power sector. Compared to an interna-

tional benchmark, the return on capital in the Icelandic power sector has been low. In the

period 1966–2010, Landsvirkjun’s average return on capital has consistently been around

3.5 per cent, see Figure 5. This is significantly lower than international benchmarks on

power generation,9 where the return on capital has been around 7.5 per cent on average.

8 It should be noted that the power prices in Figure 4 are not fully comparable, as the Icelandic prices are weighted averages

of prices on long-term contracts, some of which were signed a long time ago, while data for the other countries is rather

based on spot prices. In other words, the figure does not show what power price potential investors considering locating in

Iceland would have faced in each year, but rather the prices already existing power intensive consumers were facing. 9 Data from Australia, Ireland and the UK

0

20

40

60

80

100

2007 2008 2009 2010 2011 2012 2013 2014 2015

USD/MWh

United Kingdom Denmark Finland Sweden Norway Iceland



resource value

15

Figure 5 Low return on capital in Icelandic power sector

Note: Returns on capital in Iceland over the period 1966–2010. The international benchmark is based on

data from Australia, Ireland and the UK, with rates of return reported as ranges for each country. The

figure shows the range obtained from combining the individual country ranges.

Source: Copenhagen Economics based on Jónsson and Jóhannesson (2012)

Importantly, the 3.5 per cent return on capital for the average power plant contains both

the return on capital needed for an investor to find it attractive to invest, and the resource

rent which the Icelandic state should collect from allowing companies to harness Icelandic

natural resources. An average return of 3.5 per cent implies that even with a low resource

rent,10 the return on capital is significantly lower than similar international investment op-

portunities.11

An in-depth economic analysis has been conducted on the Káranhjúkar hydropower plant

in eastern Iceland, which was built to supply power to the aluminium smelter at Reyðarfjör-

ður. Assuming that the power contract was linked to the international aluminium price,

specified in USD, and including a right to purchase less power than the maximum stipu-

lated in the contract, the economic risks associated with the aluminium smelter and the

power plant are likely to be quite similar.12 This would imply that the required return on

capital should be similar for investors in the two facilities. Instead, the return on capital for

the smelter (9.7 per cent) was nearly double the rate for the power plant (5.3 per cent).

The fact that Iceland has had low power prices is not surprising given its abundance of

energy that can be turned into power at low cost and the lack of a connection to interna-

tional markets. However, low production cost does not necessarily mean that a producer

should sell its product at a low price. A comparison with the oil market is instructive here:

some oil fields in for example Norway produce oil at costs that are very low compared to

10 Iceland is currently levying some rent-like instruments such as licensing fees and local municipality taxes 11 Hydro plants in Norway for example pay a sizeable rent tax 12 Jóhannesson (2001). Both facilities are exposed to both price and demand fluctuations and the power plant is, in addition,

exposed to currency risk

3,5%

5,9%

9.4%

0%

2%

4%

6%

8%

10%

Icelandic

power sector

International

benchmark

Rate of return



resource value

16

the world market oil price. Yet it is not suggested that the owners of those fields should sell

at a low price, but instead at the world market price thereby capturing a substantial re-

source rent to the Norwegian community.

Higher power prices would allow Iceland to capture more resource rent from power gener-

ation. A crude illustration suggests that if the average power price for the power intensive

industry increases between USD 10/MWh and USD 18/MWh (the latter corresponding to

an increase to the current wholesale price) it would lead to a corresponding revenue gain

in the range USD 150–270 million per year, which could be channelled to the Icelandic

community. An increase to the NordPool system price would generate additional income

of around USD 130–550 million per year, depending on the development in the Nordic

power prices in 2020; see Figure 6.

Figure 6 Potential resource rent increases with higher power

prices

Note: Based on power sold to power intensive consumers only. All values are in constant (2015) USD.

'Increase to wholesale price' is based on the average price difference between the Icelandic wholesale

price and PPAs, over the period 2006-2015.

'Increase to NordPool price' is based on three scenarios for Nordic power prices in 2020, from Dansk

Energi (2016). The low-price scenario is referred to as the Forwards scenario in Dansk Energi (2016),

and has a NordPool price of around 25 €/MWh. The mid- and high-price scenarios are referred to as

the Climate and the WEO2015 scenarios, and have NordPool prices of around 34 €/MWh and 50

€/MWh, respectively. The Swedish Energy Agency (Energimyndigheten, 2014) provides a scenario for

power prices in the NordPool area, where the 2020 power price is around 42 €/MWh, i.e. halfway be-

tween Dansk Energi’s mid and high scenarios.

Source: Copenhagen Economics based on Landsvirkjun, NordPool, Dansk Energi (Danish Energy Association),

Energimyndigheten (Swedish Energy Agency) and Central Bank of Iceland

152

229

272

127

282

555

0

100

200

300

400

500

600

10 $/MWh

increase

15 $/MWh

increase

Increase to

wholesale price

Increase to

NordPool price

mUSD/year



resource value

17

Power prices can be raised without detriment to Iceland’s attractiveness as

an investment object

Iceland competes with a range of other countries for attracting power-intensive industrial

facilities. In order to remain attractive in the eyes of international power intensive consum-

ers, Iceland needs to fulfil at least two criteria:

1. Offer commercial conditions including power prices similar to other alternative lo-

cations

2. Maintain its reputation as a stable country in which to invest and do business

Power intensive consumers such as aluminium producers are operating on an international

market, where local conditions such as electricity costs are more important than the dis-

tance to the market and the associated transport costs. Both bauxite (the main ingredient

for aluminium production) and aluminium are highly tradable commodities, which do not

need to be produced close to where they are consumed. Indeed, aluminium is highly traded,

as evidenced by the high trade intensity of aluminium compared to other products, see Fig-

ure 7.

Figure 7 Trade intensity of aluminium is comparably high

Note: Trade intensity of aluminium calculated for the EU, based on trade data for 2009–2011. The trade in-

tensity is a measure of the extent to which a certain product is traded, as opposed to domestically

consumed. It indicates the tradability of a particular product type, where e.g. fresh milk has a low

trade intensity as it is very expensive to transport at lengthy distances. For a given country or region,

trade intensity is defined as (X+M)/(M+Y), where X represents exports, M imports, and Y domestic

production. See Copenhagen Economics (2011) for further details.

Source: Copenhagen Economics based on European Commission (2015), p. 172

As a result, a majority of aluminium smelters globally are located near sources of hydro-

power, to ensure cheap and reliable energy, and to waterways, to facilitate transport of in-

put materials and final output. Iceland’s location is therefore not a particular disadvantage

for aluminium smelters, and does not in itself necessitate a discount relative to power prices

in competing locations.

If power prices in Iceland become too high, however, the power intensive consumers will

relocate somewhere else, or choose not to expand production. In the current environment

of low commodity prices in general, including an aluminium price nearly 50% lower than

58%

31%25%

19%16% 14%

0%

25%

50%

75%

Basic chemicals Aluminium

production

Steel production Manufacture of

sugar

Manufacture of

cement

Manufacture of

bricks and tiles

Trade intensity



resource value

18

before the financial crisis, investment appetite is low. In order to remain attractive for such

consumers, price setting should pay close attention to the alternative international oppor-

tunities, and align power prices thereafter. Figure 8 hypothetically illustrates the room for

manoeuvre that Iceland is facing: power prices can be increased up to the point where an

alternative location becomes more attractive.13 This does not mean that Iceland has to offer

lower power prices than competing markets, but that the entire package of locating in Ice-

land should be equally good as alternatives.

Figure 8 Illustration of room for manoeuvre in price setting

Note: Hypothetical illustration of Iceland’s room for manoeuvre in price setting; not based on actual data.

Source: Copenhagen Economics

Historically, aluminium production has been the main channel through which Iceland’s

natural energy resources are turned into value. However, today the number of power con-

sumers finding Iceland attractive is higher and still increasing. Expansion of existing facil-

ities and new facilities such as data centres and silicon metal production have increased the

demand for power. There are potential sources of additional demand, which would allow

for expanding power production even further and at the same time potentially contribute

to raising the power price in Iceland, see Figure 9. This increase in demand provides a nat-

ural point for raising power prices, and indeed newer power contracts are of higher levels

than historically.

13 Note that the figure is a conceptual illustration, and not based on actual data

0

20

40

60

80

100

Iceland Competing

markets

Raw materials Labour Transport Other Electricity

Room for manoeuvre



resource value

19

Figure 9 Potential new consumers who could consider Iceland

Note: Current consumption refers to 2014 figures.

Source: Copenhagen Economics based on McKinsey (2012); Statistics Iceland (Gross consumption of electric-

ity 1990–2014)

Energy policy and environmental objectives are not sufficiently coordinated

In many circumstances, there is a trade-off between power generation and environmental

concerns. Building power plants and transmission lines changes the Icelandic landscape

and may detract both from the value Icelanders derive directly from their natural environ-

ment and from the potential for tourism. Some of this damage to the natural environment

may be irreversible in a similar manner to too much tourism. On the other hand, expanding

power generation may also develop infrastructure which can make otherwise inaccessible

land accessible.

No-build zones designed to protect the environment from the unrestricted expansion of

power generation will likely reduce the potential value generated for the Icelandic commu-

nity. There are concerns that this trade-off is not sufficiently integrated in decision-making.

While it is commendable that the Master Plan Act stipulates that assessments of power

plant expansion plans should take into account economic as well as environmental and so-

ciological considerations, in practice the assessments does not always consider all these

aspects. In particular, in the latest round of the Master Plan, economic assessments were

completely set aside. A lack of economic analysis may have been less important historically,

but it becomes more damaging as power prices increase and therefore generate more value

for the Icelandic community.

The restrictions made in the recent round of the Master Plan will increase the cost of power

from new capacity. This is because several of the most-cost efficient options have been re-

stricted, which drives the cost of new generation up. Specifically we estimate that the cost

of expanding power generation similar to the past 10 years will increase by about 84 million

USD/year from around 270 to almost 360 million USD/year, see Figure 10. This cost

should be evaluated against the potential benefit from an improved environment.

Aluminium and

basic metals

(combination of new producers and

expansion of

existing ones)

Data centres and

other industries

0

5

10

15

20

25

30

35

Current consumption Potential consumption

Electricity

consumption (TWh)



resource value

20

Figure 10 Master Plan restrictions increases power generation

costs

Note: The figure shows the cost, per year, of expanding generation by 9,400 GWh per year. This is similar

to the expansion in generation over the past 10 years, and also corresponding to the amount (and

expected utilisation) of capacity currently on the list of approved projects in the Master Plan.

Source: Copenhagen Economics based on Olafsson (2016), p. 13

358 84

274

0

100

200

300

400

Cost with

restrictions in

Master Plan

Cost of

restrictions

Cost without

restrictions

million USD/year



resource value

21

Chapter 2

2 Five options for policy reform

In order to address the potential challenges facing the Icelandic energy system, a number

of policy options could be considered. We have identified five areas, where policy reform

could be considered.

2.1 Enhancing security of supply for households and SMEs As described in Chapter 1, security of supply for households and SMEs is pressured by the

increased demand from industrial consumers as well as households and SMEs combined

with increased environmental concerns and an inflexible and relatively low price resulting

in too low incentives to invest in new capacity. In order to address this issue, the following

three options could be considered.

Improve market signals of scarcity to investors

In traditional markets, a lack of supply will increase the price providing an incentive to

expand supply. Similarly in electricity markets, new investments in capacity expansions

will be driven by increased prices. However, there is a risk that the price formation in the

market for households and SMEs does not reflect a potential underlying scarcity among

other things due to the market organisation of the retail market. Going forward, it should

be considered that prices should be allowed to rise in response to a lack of capacity, e.g. by

considering retail market reform as also suggested below in section 2.4.

Improved institutional responsibilities to safeguard security of supply

In the case of an absence of a functioning energy-only market for electricity providing a

scarcity price signal to investors, it becomes very important to ensure that a clear institu-

tional responsibility is in place to monitor and safeguard sufficient available capacity. Hav-

ing a clearly defined responsibility and mandate for ensuring security of supply, in addition

to necessary resources and competences to fulfil the role, will reduce the risk of supply

challenges.

Consider paying for a capacity reserve

Another option could be to shield the household and SME segment against price increases,

by maintaining a relatively low price and instead provide the incentives to build capacity

through other mechanisms. This could be done, e.g., by paying a capacity subsidy to any

market participant willing to supply back-up capacity, ensuring that incentives to supply

this capacity to households are at least equally attractive as supplying to the power-inten-

sive segment. It is important to recognise that continuing to serve the household market at

conditions less attractive than in the power intensive market constitutes an economic loss

to the power producers, which should be made transparent, e.g., through a subsidy pay-

ment or an accepted lower dividend payment from state owned companies back to the Ice-

landic community.



resource value

22

The viability of an option shielding the household and SME segment from price increases

should be assessed in respect of EU State Aid rules. EU State Aid rules are in place primarily

to secure an efficient internal market with equal competitive terms for all companies across

national borders. It is therefore a priori less of a concern that Icelandic households poten-

tially are getting preferential treatment than if Icelandic companies are getting preferential

treatment. Models where all companies are facing a cost-reflective power price are less

likely to be in violation of the State Aid rules. Ultimately, any model will need to be assessed

based on their concrete design.

2.2 Options to ensure fair value to the Icelandic community

from its energy resources As highlighted in Chapter 1, the value going back to the Icelandic community from power

generation in the past has been relatively low. Recent developments suggest that Iceland

has developed a quite strong position for attracting new sources of electricity demand with

a willingness-to-pay exceeding historical power contracts. This is borne out by the obser-

vation that recent renegotiations with existing power-intensive customers have resulted in

increased power prices. This puts the Icelandic community in a better position to achieve a

more favourable share of the total value generated from the power sector.

The current setup for extracting value for the Icelandic society is to a large extent left to the

discretion of Landsvirkjun, which is the largest generator in Iceland, in the absence of clear

instructions from its owner, the Icelandic state. This lack of instructions from its owners

leaves the role and objectives of Landsvirkjun unclear as it is also a commercial operator in

a competitive market. It may even inhibit the degree of competition between energy pro-

ducers if Landsvirkjun’s return on equity is lower than the required return on equity of

commercially owned producers.

Instead, the Icelandic community should consider defining its expected returns from power

generation. The returns should be as high as possible while still rendering Iceland an at-

tractive location for electricity-intensive consumers. An anchor for this could be to aim at

setting power prices close to relevant international benchmark prices in order to mimic the

alternative opportunities for electricity intensive consumers.

Two options to consider would be 1) setting up a transparent rent tax system to ensure that

the Icelandic community would reap a fair value from its natural resources, and 2) defining

a required return on equity for state owned companies in excess of the rent tax, thereby

ensuring competition on equal terms with commercial entities.

Setting up a transparent rent tax system

A transparent rent tax system would help identify the value the Icelandic community is

expecting to get in return for allocating the right to harness power from Icelandic natural

resources. Ideally, this would replace some existing rent-like taxes, such as property taxes

on power installations and water and geothermal rights payments, which are currently ad-

ministered at the local level, with a transparent compensation to local municipalities e.g.

from the State. This would, in turn, ensure equal conditions for all companies, whether



resource value

23

state owned or private, and all locations, and provide clarity about the take needed to

achieve an acceptable return on equity (for private and public owners) and the take related

to the company’s access to valuable natural resources.

The specific design of the rent tax is important and several different variations of a rent tax

can be found in different countries and between different sectors such as hydrocarbon ex-

traction, mineral mining, fisheries and electricity generation. It is important that it is de-

signed to provide the best incentives in the specific Icelandic context. A few examples of

elements in the rent tax which are known from other countries could be:

A one-off concession/license payment: makes government-take very secure and

moves a high degree of risk to the licensing party in terms of future revenue

streams.

A royalty payment based on revenue: commonly used in the mineral mining indus-

try, simple to administer and splits risk to future revenue streams between the gov-

ernment and the licensing party. It places all risk concerning cost developments in

the hands of the licensing party.

A tax element linked to a benchmark such as, e.g., the Nordpool price, i.e. the tax

increases when the Nordpool price increases and vice versa. This would automati-

cally link government-take to the degree of attractiveness Iceland has vis-à-vis

other locations for power-intensive consumers.

A tax element linked to actual profits, taking into account the fact that the costs of

the underlying natural resources is different in different locations in Iceland.

Defining a required return on equity from state-owned companies

By defining a required rate of return in the form of dividend payments from state-owned

companies, this provides a clear signal about the level of profitability expected when state-

owned companies build and operate power facilities. This will contribute to Iceland’s state-

owned companies operating under equal conditions with privately owned companies and

reduce the risk that some contracts are concluded at power prices that would constitute

implicit state aid.

Making value to the Icelandic community from power generation more

transparent

Options to increase transparency of the flow of value from power generation to the Ice-

landic community should be considered. While 'earmarking' government revenue for par-

ticular purposes does not generally make good economic sense, it would help mitigate the

lack of public support for raising power prices to power-intensive consumers even if it

might mean also raising power prices to households and SMEs. Options to consider could

be to channel the revenue from a rent tax or the dividend payments from state-owned

power companies into a sovereign fund, or simply reducing the general tax level in the

country through reduced labour taxes, or paying out a yearly ‘resource rent cheque’.



resource value

24

2.3 Integrating value from environment and energy Harnessing power from natural resources can often be in opposition to enjoying the envi-

ronmental value of the same resource. In Iceland, this trade-off is supposed to be assessed

in the Master Plan act, where specific energy projects are deemed to be in accordance or

not with environmental objectives. This type of framework is commendable, as it has the

potential to make an integrated assessment taking into account multiple objectives.

However, concerns have been raised that the practical assessment has not sufficiently taken

into account the direct and indirect value from energy generation, thereby risking subopti-

mal outcomes for the Icelandic community. Specifically, when restricting particular zones

or power expansion projects, the economic cost of doing so in terms of increased cost of

energy generation or even the leading to ‘too little energy production’ should be taken into

account. This becomes increasingly important if power prices are increased, as power gen-

eration will then generate a higher value to the Icelandic community. The cost to Iceland

from these restrictions will then also increase.

2.4 Market organisation in the Icelandic energy markets Energy generation in Iceland is concentrated among few companies. Three companies gen-

erate 97 per cent of all electricity. Out of this, Landsvirkjun is the largest player, constitut-

ing about 73 per cent.14 This is not a unique situation compared to other countries where in

14 out of 24 EU countries, the incumbent electricity generator holds a market share of more

than 50 per cent, and in seven countries the market share is above 70 per cent, see Table 1.

Table 1 Market share of the largest power generator across Eu-

ropean countries Below 30% 30-50% 50-70% Above 70%

Number of countries

6 4 7 7

Countries ES, IT, LT,

PL, RO, FI DE, DK, PT, SE

BE, CZ, HU,

IE, LV, LU, SV

CY, EE, FR, GR,

HR, MT, SK

Note: 2014 data. Austria, Bulgaria, the Netherlands and the UK are not included due to lack of data.

Source: Copenhagen Economics based on Eurostat (ten00119)

In most industries, ‘economies of scale’ is important. This is especially so, if there are im-

portant back-bone tasks that can be leveraged in bigger companies, and would inefficiently

be duplicated by having too many companies. There are reasons to believe that such scale

effect are significant in power generation. In fact, in all Nordic countries, Germany, France

and the UK (among many others) the largest energy company is significantly larger in terms

of energy generated than Landsvirkjun. Except for Danish DONG, which is similar in size,

the Nordic companies Statkraft, Fortum and Vattenfall produce 5-15 times more energy

than Landsvirkjun, see Table 2. Indeed, larger players such as EDF and E.ON produce more

than 50 times more energy than Landsvirkjun.

14 If sales to power-intensive industry is excluded, Landsvirkjun’s market share is approximately 55%.



resource value

25

Table 2 Landsvirkjun’s generation is low in international com-

parison

Country Largest

generator

Total domestic

generation

(TWh)

Market share of

largest genera-

tor

Estimated do-

mestic genera-

tion of largest

generator (TWh)

Total genera-

tion of largest

generator

(TWh)

Iceland Landsvirkjun 18 73% 13 13

Denmark DONG Energy 31 37% 11 14

Norway Statkraft 142 31% 43 56

Sweden Vattenfall 151 43% 65 173

Finland Fortum 65 25% 16 73

France EDF 541 87% 470 624

Germany E.ON 562 32% 180 215

Great Britain EDF 318 29% 93 624

Note: Data for 2013 (Great Britain) and 2014 (all other countries)

Source: ENTSO-E, Eurostat (ten00119), Annual reports of Landsvirkjun, Dong, Statkraft, Vattenfall, Fortum,

EDF and E.ON

Even so, there is cause to reflect on whether this kind of market organisation is appropriate.

When market concentration is high, it is natural to be concerned about inefficient markets,

e.g., in terms of setting prices too high. This is traditionally assessed in terms of market

efficiency and operational efficiency respectively.

Market efficiency is reduced if a monopoly can restrict production in order to gain a higher

price. This is inefficient as some transactions which would have been valuable to society are

not taking place (a deadweight loss).

Operational efficiency is reduced if a monopoly is not facing sufficient competitive pres-

sure allowing it to make inefficient decisions. This may lead to suboptimal investments, so

called goldplating of, e.g., investments, overly high wages and bonuses, and a lack of inno-

vation.

Based on our analysis, we argue that the case for reorganising the market for energy gener-

ation is not particularly strong. However, a potential for a lack of operational efficiency

should warrant attention. We base this on particularly two arguments: 1) substantial parts

of the electricity generation is supplied to a market subject to strong international compe-

tition and 2) potential challenges in Iceland stem from power prices being too low, not too

high due to monopoly pricing.

Instead, there may be merit in considering a reorganisation of the market for supplying to

households and SMEs, which is very different in nature than the market for supplying to

power intensive consumers:

Market for power-intensive consumers

In this market there are currently three main companies: Landsvirkjun supplying about 77

per cent, Orka Náttúrunnar supplying 17 per cent, and HS Orka supplying 5 per cent. Even

though this market is very concentrated in terms of suppliers, it is highly competitive in



resource value

26

nature. Large global corporations producing commodity products are deciding to locate

production facilities based on a global menu of location options. If power prices in Iceland

were to increase substantially above other locations, Iceland would reduce/lose its attrac-

tiveness as an investment object. This sets a natural ceiling for the degree of price increases

possible by the generators.

It is also important to recognise that high prices do not necessarily imply inefficient mar-

kets. If Landsvirkjun continues to supply the same amount of power but for a higher price,

this does not give rise to any inefficiencies. Instead it will lead to a transfer from power

consumers to power producers (primarily the Icelandic state). There is reason to believe

that prices indeed could be raised without reducing the electricity demand. As discussed in

Chapter 1, new power-intensive consumers seem available at prices significantly higher

than the historically low levels.

In addition, while Landsvirkjun is the dominant player in the market, there is de facto com-

petition between the energy companies. HS Orka, Orka Náttúrunnar and potentially other

companies can also offer contracts to power-intensive consumers, which has indeed also

taken place. The main challenge with respect to market entry is likely to be environmental

regulations restricting the building of new power plants.

A potential concern is whether Landsvirkjun’s dominant position restricts new potential

generation capacity from being built thereby restricting available capacity in the market

going forward. This would be monopoly behaviour, which would contribute to ‘too high’

prices and a ‘too high’ return on capital from its assets. However, the situation in Iceland is

in fact the opposite. Power prices have been so low that returns on capital on power plants

historically have been way below international benchmarks (as shown in Chapter 1). The

problem is therefore more about prices being too low to make new capacity commercially

viable than a deliberate restriction in capacity to keep prices artificially high.

The main drawback from the high market concentration in Iceland is the lack of commer-

cial pressure on Landsvirkjun’s decisions with the potential for inefficient operation in fu-

ture, overly high salaries etc. We do not have any indications of this taking place, but it is a

risk. To address such a risk, the natural option to consider would be to use governance

structures, e.g., commercially determined required rates of return, and performance-based

contracts, as also touched upon in policy area 1.

Market for households and small- medium sized businesses

Landsvirkjun generates about 55 per cent of this market’s total supply, while six other com-

panies generate the remaining 45 per cent. These six companies are also active in the retail

market of supplying energy to consumers. The current business model is that the six inte-

grated companies generate energy to their retailers’ consumers, and buy the residual en-

ergy from Landsvirkjun. This implies that Landsvirkjun is the only seller on the wholesale

market, and its prices thereby sets the market prices.

Due to this, there have previously been suggestions about making the wholesale market

more ‘market-based’. A key ingredient in obtaining a more market-based system would be



resource value

27

to disrupt the close relationship between the local energy companies’ generation portfolio

and its retail consumers. This could be achieved by establishing a market design where the

companies would find it commercially beneficial to offer its generation to a market instead

of directly to its retail arm, or through direct regulation e.g. by requiring that a certain share

of generation needs to be available for a general market or alternatively full ownership un-

bundling of the generation, and distribution and retail parts of the energy companies. Sev-

eral options could be envisaged for obtaining a more market-based system. Here we touch

upon two, namely 1) a full-fledged exchange-trade solution, and 2) an ‘energy pool’ solu-

tion:

A fully-fledged exchange-trade solution, e.g., by establishing a power exchange, would

have a number of interesting properties. These include high transparency in the market

thereby lowering search and transaction costs for market participants, and giving rise to a

price reflecting the true valuation of power. Conversely, it requires a significant organisa-

tional setup to function properly and the establishment of a well-functioning financial mar-

ket as well. The Icelandic retail market is quite small, and a full-fledged exchange solution

might struggle to provide benefits in excess of the costs of administration. For similar rea-

sons, in fact very few small, isolated regions have adopted power exchanges. Besides from

Tasmania which joined the Australian power exchange after connecting to the mainland,

and New Zealand which is significantly larger than Iceland both in terms of population and

energy consumption, most other small, isolated regions trade electricity without an ex-

change, including Alaska, Canary Islands, Cyprus, Hawaii and Malta, see Table 3.



resource value

28

Table 3 Very few small isolated power markets with exchange-

based trade

Population (million)

Electricity

consumption

(TWh)

Electricity

consumption

(MWh/capita)

Isolated system

Power exchange

Iceland 0.33 17.7 53.8 Yes No

Alaska 0.74 6.3 8.5 Yes No

Canary

Islands 2.13 8.6 4.0 Yes No

Cyprus 1.17 4.1 3.5 Yes No

Hawaii 1.43 9.5 6.6 Yes No

Malta 0.43 2.0 4.7

Until March 2015;

since then

connected to Sicily

No

New Zealand 4.6 40.4 8.8 Yes

Yes

(New Zealand

Electricity Market)

Tasmania 0.52 9.8 18.8

Until December 2005,

since then connected

to mainland Australia

Yes

(Australian National

Electricity Market)

Note: Population figures from 2015; electricity consumption from 2013.

Source: World Bank, IEA, Orkustofnun, EIA, US Census Bureau, Executive Agency for Competitiveness and

Innovation (2010), New Zealand Electricity Authority, Australian Bureau of Statistics, Office of the

Tasmanian Economic Regulator (2016)

If consumption from power intensive industry were also to become exchange traded, the

business case for an exchange would improve. However, the large industrial consumers are

very interested in stability of price and delivery conditions, for which individual power pur-

chasing agreements (PPAs) are well suited.15 This makes it unlikely that the large industrial

consumers will want to move their business to an exchange unless the same certainty can

be established there. In addition, a market based almost purely on energy sources with zero

(or very low) marginal costs may prove difficult to run without massive price fluctuations

and unpredictability.

Another option could be a model inspired by the so called ‘Energy pool’ adopted by the UK

in 1990, see Box 1. The main idea behind such a pool is to introduce an independent buying

agent on behalf of all retail consumers in the household and SME segment. This agent

would thus be responsible to purchase sufficient amount of energy from generators e.g.

through auctions and/or tender processes. Such a model would likely achieve important

objectives of 1) securing energy at the lowest possible price for consumers by buying from

15 Historically PPAs have been issued in USD and linked to global commodity prices alleviating both currency market and

commodity market risks from the power intensive consumers such as aluminium smelters. OECD estimated in 2015 that 50

per cent of power contracts were linked to global aluminium prices (OECD (2015), page 26.)



resource value

29

the most efficient generators, and 2) improving security of supply for this segment by plac-

ing a responsibility for this on the buying agent thereby also giving an incentive to include

scarcity concerns in the power price, if supply of capacity is deemed to be too low. Such a

model would require economic regulation of the buying agent in order to determine a price

then charged to retail consumers in line with a proper measure of costs of running the en-

ergy pool. While the UK version established a daily day-ahead spot market, the market in

an Icelandic context could probably function well with fewer auctions.

Box 1 The Electricity Pool of England and Wales

The Electricity Pool of England and Wales was a market mechanism in place between

1990 and 2001. It was an unincorporated association of its members, wholesale buyers

and sellers of electricity. The pool was operated and its settlement system was admin-

istered by the National Grid Company (NGC), the TSO. The House of Commons Trade

and Industry Committee noted that the purpose of the Pool had not been set out, but

they understood its three main functions to be determining the merit order, determining

the prices for services traded, and ensuring sufficient capacity to maintain the system

security.

The pool was a compulsory, day-ahead, last-price auction. The NGC provided an esti-

mate of day-ahead system demand, and required generators to submit bids to the day-

ahead auction. Based on these bids, the NGC calculated a generation schedule, and

determined the System Marginal Price (SMP) based on the highest-price bid required to

meet demand. Generators were paid the SMP, plus a capacity payment.

Retail price increases were subject to price caps linked to the retail price index, and to

cost increases such as administration payments to the pool.

The Electricity Pool suffered from a number of problems, most importantly that it created

an effective duopoly, where the two main generators set the price over 90 per cent of

the time. Other criticisms were that it was only half a market with inadequate represen-

tation of the demand side, that was opaque, unpredictable, and therefore hard to hedge

using standard contracts, and that was compulsory which prevented trading outside the

Pool and hence discouraged contracting. To some extent, a number of these issues

would likely be less severe in an Icelandic context.

Source: Copenhagen Economics based on Simmonds (2002); Newbery (1998)

A more market-based pricing is a prerequisite for flexible pricing, which can reflect varia-

tions in demand and supply, e.g., over the day or across seasons. This would be important

in order to activate potentially flexible demand both in the retail and SME segment but also

in the power intensive segment, if there is a business case for this. Another requirement is

installation of smart metering devices. The value to the Icelandic community of pursuing

such price flexibility is very dependent on variations in demand and supply, which, e.g., will

increase if the share of electric vehicles increases.

Consider separating energy companies’ ownership of Landsnet

The Icelandic TSO Landsnet is currently owned by the previous owners of the transmission

grid, i.e. the energy companies, in particular Landsvirkjun. This may create improper in-



resource value

30

centives in the choices of Landsnet, e.g. related to investments in new transmission capac-

ity, including connections to new power plants. This is not ideal, and a plan for divestment

could be considered.

2.5 Interconnector project to be considered as an option While Iceland's power system is currently isolated, the possibility to build a connection to,

e.g., the UK has been discussed. This could be an attractive investment for Iceland for a

number of reasons. It would raise the value of Icelandic electricity production as the UK is

an area with high power prices. In addition, Iceland's large hydropower capacity provides

plenty of balancing possibilities, which would be valuable to the UK market, which has a lot

of intermittent power sources. An interconnector will also improve security of supply in

Iceland, both through the physical connection, which allows for electricity imports when

needed, and as a result of creating a new market place where power is sold at the UK price,

thereby sending a price signal to new potential investors in new generation in Iceland.

An interconnector is a long-term option, however, as lead times on decisions and subse-

quent construction are substantial. In the meantime, it is important that any policy reform

be designed with this option in mind, i.e. by not locking Iceland into structures that will

prove sub-optimal in the case that an interconnector is built. One example of this would be

for Iceland to avoid entering into very long-term power contracts - unless they are written

in a way that is flexible with respect to the option of building an interconnector, e.g., by

allowing the price to be linked to the market in the event that an interconnector is built.



resource value

31

References

Christensen, L., Our Energy 2030 – Efficiency, com-

petitiveness and transparency in the Icelandic en-

ergy sector, Markets & Money Advisory, Copen-

hagen

Copenhagen Economics (2011), Carbon Leakage

from a Nordic Perspective

Dansk Energi (2016), Elprisscenarier 2020–2035

Energimyndigheten (2014), Scenarier över Sveriges

energisystem – 2014 års långsiktiga scenarier, ett

underlag till klimatrapporteringen

European Commission (2015), Impact Assessment,

accompanying the document “Proposal for a Di-

rective of the European Parliament and of the

Council amending Directive 2003/87/EC to en-

hance cost-effective emission reductions and low-

carbon investments”

Executive Agency for Competitiveness and Innova-

tion (STORIES project) (2010), Scheme for mar-

ket organization of autonomous electricity sys-

tems

Icelandic Parliament (2003), Electricity Act, No.

65/2003

Jóhannesson, S (2001), Látum markaðinn meta Ká-

rahnjúkavirkjun. Vísbending, s. 2, 14 December

2001

Jónsson, Á., and Jóhannesson, S (2012), Arðsemi

Landsvirkjunar af rafmagnssölu til stóriðju. Uni-

versity of Iceland, Institute of Economic Studies,

Reykjavik.

Lazard (2015), Levelized Cost of Storage Analysis -

Version 1.0.



resource value

32

McKinsey & Company (2012), Charting a Growth

Path for Iceland

Newbery, D. M. (1998), The Regulator’s Review of

the English Electricity Pool

OECD (2015), OECD Economic Surveys: Iceland

2015, OECD Publishing, Paris

Office of the Tasmanian Economic Regulator (2016),

Energy in Tasmania - Performance report 2014-

2015

Ólafsson, K. B. (2016), Levelized Cost of Energy og

virkjunakostir til umfjöllunar í 3. áfganga

rammaáætlunar

Orkustofnun (2015), Raforkuspá 2015–2050

Simmonds, G. (2002), Regulation of the UK Electric-

ity Industry, CRI Industry Brief, University of

Bath School of Management

Documents

Energy market reform options in Iceland - Copenhagen Economics · Policy area 3: Integrated assessment of energy and environmental objectives Harnessing power from natural resources